We finally have an update for...PNG?
Posted Jun 25, 2025 at 1:50 PM UTC We finally have an update for...PNG?
Launching 22 years after its last major update, the latest PNG spec now includes native support for HDR, APNG animations, and Exif metadata for embedding information into image files. W3C PNG Working Group chair Chris Blume says Chrome, Safari, Firefox, iOS, macOS, and Adobe Photoshop already support the new standard, and that upcoming updates will improve compression and dynamic range support. PNG is back!
[programmax.net]
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
TechCrunch
23 minutes ago
- TechCrunch
Meta is adding AI-powered summaries to WhatsApp
Meta announced on Wednesday that it's adding an AI-powered summaries feature to WhatsApp. The optional new feature uses Meta AI to summarize unread messages in a chat. This summary would only be visible to you, not others in your chat, the company notes. The feature builds on the AI technology that Meta released in April, which allowed the company to implement AI features that don't impact encryption or user privacy. 'Message Summaries uses Private Processing technology, which allows Meta AI to generate a response without Meta or WhatsApp ever seeing your messages or the private summaries. No one else in the chat can see that you summarized unread messages either,' the company said in a blog post. Image Credits: Meta Meta is initially rolling out the feature in the U.S. with English language support. It will reach more countries and languages later this year. Until now, users could access Meta AI within the chat to ask general questions or tag a message to give the chatbot context. However, Meta AI couldn't read your messages. Meta said that the new stack allows WhatsApp to access context from your chat privately to process requests through its AI. This allows it to summarize messages or provide writing suggestions. Techcrunch event Save $200+ on your TechCrunch All Stage pass Build smarter. Scale faster. Connect deeper. Join visionaries from Precursor Ventures, NEA, Index Ventures, Underscore VC, and beyond for a day packed with strategies, workshops, and meaningful connections. Save $200+ on your TechCrunch All Stage pass Build smarter. Scale faster. Connect deeper. Join visionaries from Precursor Ventures, NEA, Index Ventures, Underscore VC, and beyond for a day packed with strategies, workshops, and meaningful connections. Boston, MA | REGISTER NOW The new AI-powered features are available under a new setting that can be accessed via Settings > Chats > Private Processing, which lets you turn on or off individual functions.
Forbes
an hour ago
- Forbes
The Future Of Agentic AI Isn't Just Autonomous—It's Accessible Anywhere
Shahar Chen, CEO and Cofounder of Aquant, a company focused on delivering generative AI solutions to transform the service industry. Everyone's talking about AI agents. But in all the noise, one thing is clear: Most of the conversation is still stuck in the theoretical. What does agentic AI really mean for those of us solving real-world problems, often in high-stakes environments, with legacy infrastructure and frequently without reliable connectivity? The next wave of agentic AI won't just be about more autonomy. It'll be about more accessibility, more resilience and more relevance. Agentic AI In The Real World: Less Gloss, More Grit Let's set the record straight: Agentic AI isn't just a flashy new product category. It's an intelligence layer—one that enables software to reason, adapt and plan. It powers everything from decision support to autonomous workflows. But in industries like manufacturing, healthcare, logistics, utilities, telecommunications and aviation, AI can't live in a bubble. These aren't controlled environments with always-on internet. These are factory floors, hospital wings, shipping yards, cell towers, wind farms and airline maintenance hangars—places where uptime matters and decisions have consequences. That's why agentic AI needs to be more than 'smart.' It needs to work offline without fail. Offline AI: The Unsung Hero Of Agentic Intelligence Here's a truth that's often overlooked: The most advanced AI is useless if it disappears the moment the signal drops. Consider the technician maintaining aircraft avionics on a remote runway or the utility worker restoring power after a storm. How about the nurse in a rural clinic accessing treatment guidelines or the warehouse manager resolving an inventory issue in a dead zone of connectivity? In each case, critical decisions can't wait for a network to reconnect. That's where offline AI becomes indispensable—delivering intelligence and support right where it's needed, with no signal required. Smarter Doesn't Mean More Autonomous—It Means More Aware There's a growing buzz around AI agents—task-completing tools designed to follow instructions, automate workflows and take action with minimal human input. But agentic AI is something different and more powerful. Agentic AI goes beyond agents. It's not just about executing commands—it's about reasoning through decisions, adapting to changing conditions and acting with awareness of goals, context and constraints. It's AI with a sense of responsibility, not just automation. In high-stakes environments, the goal isn't full autonomy for autonomy's sake. The goal is intelligence. And that means: • Knowing when to act and when to escalate • Making decisions with incomplete or conflicting data • Offering guidance without overstepping boundaries • Delivering value, even when disconnected from the network The smartest AI doesn't just do; it understands. It knows when to step forward, when to hand off and how to support humans in the moment. And it's not about replacing people. It's about reinforcing them with AI that's reliable, responsive and always aware of its role. Innovating With Offline Capabilities In Mind So, what can tech leaders do with all of this? If you're building products or deploying AI in high-stakes, low-connectivity environments, offline capability isn't a 'nice to have.' It's foundational. But that doesn't mean you need to build it all yourself—in fact, you probably shouldn't. The fastest path to impact isn't building a custom offline agentic AI solution. It's investing in a platform that's already done the hard work—one that's been tested in the field, built to run in real-world conditions and designed with the edge in mind. Here are a few things to consider when evaluating platforms: 1. Look for solutions purpose-built for your industry. Generic AI tools often fall short in high-stakes environments. Prioritize platforms trained on domain-specific data and tuned for your workflows, not just general-purpose chatbots with a new label. 2. Demand offline-by-design, not just offline-capable. There's a difference between a tool that can work offline and one that was built to. Ask how the platform handles connectivity loss, what data it caches locally and how it syncs once back online. 3. Ask about field-tested deployments. Has the platform been used in environments like yours (e.g., factory floors, remote clinics or wind farms)? Real-world resilience can't be faked in a demo. 4. Think beyond the AI model. Offline agentic AI isn't just about the model—it's about the ecosystem. That includes lightweight interfaces, smart data sync, embedded escalation paths and the ability to run on limited hardware. 5. Avoid the DIY trap. Building your own system may sound appealing, but stitching together LLMs, offline logic, device compatibility and user-friendly UX is a massive, ongoing lift. The hidden costs add up fast. Investing in a platform that gives you a running start allows your teams to focus on what matters most: delivering outcomes. Because when the signal drops, your AI still needs to show up. What Comes Next: Grounded, Gritty And Built For The Field The future of AI isn't floating in the cloud—it's embedded at the edge. It's climbing utility poles, navigating hospital basements, sailing on cargo ships and inspecting infrastructure in remote areas. It's serving humans, not just in ideal conditions but in real conditions. Because in the real world, the best AI doesn't just act; it shows up. Even when the Wi-Fi doesn't. Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
Forbes
an hour ago
- Forbes
20 Strategies To Unlock AI's Hidden Potential In Fintech Operations
For fintech leaders, AI can do more than power customer-facing tools—it can quietly transform how the entire business operates. From automating risk assessments to simulating customer journeys, AI can streamline processes that drive both efficiency and insight. The potential internal and external benefits are many, including faster decision-making, leaner teams, fewer manual errors, stronger service and improved security. Below, members of Forbes Business Council share high-impact, often overlooked ways to apply AI across your fintech operations. Whether you're looking to optimize back-end workflows, enhance fraud detection or surface real-time financial insights, their strategies offer actionable starting points. 1. Deliver Hyperpersonalized Services AI empowers fintech leaders to achieve unprecedented efficiency through real-time microsegmentation and hyperpersonalized workflows. Internally, it auto-configures fees, tax withholding and payouts, cutting manual tasks. Externally, customers access custom dashboards, tailored cashflow analytics and on-demand embedded finance, driving engagement and revenue growth. - Jack Thorogood, Native Teams 2. Collaborate Securely With Federated Learning Models Leverage federated learning to train machine learning models collaboratively across institutions without sharing raw data. This boosts fraud detection accuracy, refines credit scoring, enhances client trust via data privacy, and fosters financial inclusion through secure, collective intelligence. - Juan Arroyo, SG CONSULTING GROUP Forbes Business Council is the foremost growth and networking organization for business owners and leaders. Do I qualify? 3. Tackle Fraud And Enhance CX When one of our old fintech clients approached us with the problem statement to tackle fraud and enhance CX, we implemented AI for real-time anomaly detection and user behavior-based personalization. Internally, it saved manual work and enhanced decision-making. Externally, it resulted in a significant fraud decrease, enhanced support and greater ROI through trust, retention and customer satisfaction. - Naresh Prajapati, Azilen Technologies 4. Reduce Support Desk Messages AI is a great help desk tool. When you have customers logged into your software, you can have an AI bot that can answer questions, which will cut down on phone calls and emails to your support desk. The AI bot can direct them to the answer. We did this with one of our products, and it cut calls by 40%. It makes it easy for users to get answers without having to scroll through pages. - Allen Kopelman, Nationwide Payment Systems Inc. 5. Streamline Underwriting With Automated Risk Assessment Fintech leaders can use AI to streamline underwriting and automate risk assessment through cash flow and transaction data. Internally, this reduces processing time and manual error by a significant margin. Externally, it strengthens our industry's ability to assess and unlock financial opportunities for underbanked communities. The result is improved growth, access and financial inclusion. - Raj Tulshan, Loan Mantra 6. Automate Complex Services To Reduce Cost And Error AI technologies in fintech include solutions for such activities as credit assessment, fraud detection and personalized advice. One idea to improve operations is the automation of complex financial services that previously required qualified personnel. AI allows us to provide the above services in a peopleless mode, cutting internal expenses while decreasing errors for external customers. - Anton Alikov, Arctic Ventures 7. Assess Customer Risk In Real Time Fintech leaders can use AI to dynamically assess customer risk profiles in real time using behavioral and transactional data, not just credit scores. Internally, this streamlines underwriting and reduces fraud; externally, it enables faster approvals and more inclusive financial products. - David Centeno, David Centeno Law, PC - NY Divorce Lawyer 8. Predict Customer Financial Stress Before It Erupts One unique way fintech leaders can leverage AI is by using predictive analytics to identify customer financial stress before it escalates. AI can flag patterns like unusual spending, missed payments or sudden cash flow changes. Internally, this enables proactive customer support and risk management. Externally, it builds trust, improves retention and positions the company as a true financial partner. - Stephen Sokoler, Journey 9. Continuously Authenticate Users One way fintech leaders can leverage AI is through real-time behavioral authentication that continuously verifies user identity based on typing patterns, mouse movements, device interaction habits and transaction timing. The technology becomes smarter, learning to distinguish between legitimate changes in user behavior versus potential security threats, creating a more adaptive security layer. - Jay Mehta, Seldon Capital 10. Stress Test Your Business With Crisis Simulations Fintech leaders should deploy AI algorithmic stress testing, simulating thousands of crisis scenarios—from cyberattacks to market crashes—to identify vulnerabilities impossible to spot manually. Internally, this creates unprecedented resilience and optimizes resources. Externally, it builds stakeholder confidence and ensures service continuity when customers need financial services most. - Oleg Levitas, Pravda SEO Inc., Real Results SEO Inc. 11. Simulate Customer Journeys With Digital Twins Use AI-driven digital twins of customer journeys (including onboarding, transactions and so on) to simulate and optimize workflows before launch. Internally, teams can pinpoint bottlenecks, slash cycle times and tailor segments. Externally, clients get frictionless onboarding, instant approvals and hyper-personalized services, boosting loyalty and market agility. - Arpit Jain, SEO Sets 12. Accelerate KYC/KYB Verification AI-driven Know Your Customer and Know Your Business verification can make things much easier for customer service teams in fintech companies. By automating the verification process, AI can quickly flag suspicious activity, reducing the workload for CS teams. This means they can focus more on resolving complex issues instead of handling routine verifications. It leads to faster response, fewer errors and a smoother experience. - Sabeer Nelliparamban, Tyler Petroleum Inc. 13. Adapt To Users' Needs In Real Time Fintech leaders can harness AI to build real-time behavioral analytics engines that adapt to individual user actions. This enables dynamic risk profiling and personalized product delivery. Internally, it enhances efficiency and accuracy; externally, it fosters trust, deepens engagement and broadens access, especially for underbanked communities, positioning the firm as a forward-thinking financial partner. - Richard Powell, APC Holdings, LLC 14. Mine Customer Conversations For Actionable Insights Fintech leaders can use AI to mine customer conversations—calls, chats and tickets—for patterns in objections, sentiment and language. Internally, this sharpens the product and messaging. Externally, it means better targeting, faster sales cycles and stronger client trust. It's insight, not just automation, that gives you the edge. - Henry McIntosh, Twenty One Twelve Marketing 15. Identify Customer Hesitation Early We use AI to detect early signs of customer hesitation by analyzing real-time behavioral patterns. Internally, this lets us act proactively to reduce churn and improve product decisions. Externally, it helps customers feel genuinely understood, building trust in a tough industry. We see AI not just as a tool, but as a key driver of growth and lasting loyalty. - Sahil Gandhi, Blushush 16. Automate Treasury Management An AI-driven treasury engine can predict intraday inflows and outflows across chains, then auto-rebalance liquidity and hedge exposure in real time. Internally, this frees ops teams from manual transfers, trims idle capital and flags anomalies before they hit the books. Externally, counterparties get faster settlements and steadier spreads, boosting confidence without extra headcount. - Alvin Kan, Bitget Wallet 17. Detect Edge Cases Use AI to detect edge cases, not just patterns. Most risk hides in the anomalies that rules miss. Train models to flag the weird, not just automate the known. Internally, this reduces blind spots. Externally, it builds trust by catching what others overlook. - Yves Remmler, Endeavor Elements, Inc. 18. Simplify Payments Via Text Reminders One of the best ways for fintech to use AI to improve business operations is to text customers with loan payment reminders or account information, such as low balance alerts. Those businesses that allow customers to pay on their loans by clicking a text to approve payment with a card on file will gain more Gen-Z customers who love the simplicity. - Baruch Labunski, Rank Secure 19. Boost Marketing Efficiency With Generative AI While certain regulations in fintech make the use of AI and machine learning difficult to apply, business leaders in that space can leverage AI for their sales and marketing needs. AI can help brainstorm social media content buckets, draft website content and even write email copy, allowing business owners to save valuable time and focus efforts where they're most needed. - Emily Reynolds, R Public Relations Firm 20. Augment Human Capabilities And Amplify Customer Intent Fintech leaders should deploy AI that enhances human capabilities without replacing them. By designing technology that amplifies customer intent rather than eliminating choice, companies create more meaningful experiences. This improves operational efficiency internally while externally demonstrating respect for customers through transparency and giving them meaningful control. - Talbott Roche, Blackhawk Network
